Method for assigning particular classes of interest within measurement data
A method and system for surveying and/or metrology for assigning particular classes of interest within measurement data, wherein an assignment of at least one measurement object to a first class of interest within the measurement data based on a classification model, is processed by a feedback procedure providing feedback data for a training procedure which provides update information for the classification model, wherein the training procedure is based on a machine learning algorithm, e.g. relying on deep learning for supervised learning and/or unsupervised learning.
1. A system for surveying for assigning a class of interest within measurement data of a total station or a laser tracker, the system comprising:
a measurement device, selected from the total station or the laser tracker, with a data acquisition unit for acquiring measurement data for at least one measurement object having a reflector with specific properties, and
a classification unit for executing a classification of the measurement data by a classification algorithm for providing an assignment of the at least one measurement object to a first class of interest within the measurement data, based on:
the measurement data, and
an classification model made available to the classification unit comprising at least the first class of interest,
a feedback functionality for providing, in parallel with the acquisition and classification of the measurement data, feedback data based on user feedback by processing the assignment by a feedback procedure comprising at least one of:
verification information for the assignment of the measurement object to the first class of interest,
change information of the assignment of the measurement object to the first class of interest,
a definition of a new class of interest,
an instruction for removal of a class of interest from the classification model,
a first selection of the measurement data to be used for the classification, and
identification of a second selection of the measurement data to be ignored for further processing,
wherein the feedback data is provided to a training unit comprising a training procedure, the training procedure being based on a machine learning algorithm and providing update information for the classification model, wherein the training procedure is adapted for processing feedback data from a multitude of feedback procedures taking into account at least one of a multitude of different objects, time data, and position data.
2. The system according to claim 1 , wherein the feedback data being provided is based on at least one of:
explicit feedback by means of the feedback functionality for the acquisition of the measurement data or by a software used for processing of the measurement data, and
implicit feedback.
3. The system according to claim 1 , wherein the feedback functionality is supported by a notification functionality providing a status information of the classification, and wherein the notification functionality is based on at least one of:
a visual representation of the status information,
an acoustic notification of the status information.
4. The system according to claim 1 , wherein the training procedure is adapted for at least one of:
processing feedback data provided by classifications of a multitude of measurement data, and
processing additional data provided to the training procedure.
5. The system according to claim 1 , wherein the update information for the classification model is made available to a multitude of classification algorithms.
6. The system according to claim 1 , wherein the classification model is provided out of a set of different models, differing from each other by at least one of:
a region-specific classification parameter,
a region-specific class of interest,
a time-specific classification parameter,
a time-specific class of interest,
an application-specific classification parameter, and
an application-specific class of interest,
wherein the classification model is provided based on at least one of a time information, a location information, and an application information, corresponding with the acquisition of the measurement data.
7. The system according to claim 1 , wherein the classification is based on at least one of:
a class of interest based on a semantic property,
a class of interest based on a geometrical property,
linear classification,
a support vector machine,
a quadratic classifier,
Kernel estimation,
boosting,
a decision tree,
deep learning, and
learning vector quantization.
8. A system for surveying for assigning a class of interest within measurement data, the system comprising:
a measurement device with a data acquisition unit for acquiring measurement data for at least one measurement object, which is at least partly captured by the measurement data, and
a classification unit for executing a classification of the measurement data by a classification algorithm for providing an assignment of the at least one measurement object to a first class of interest within the measurement data, based on:
the measurement data, and
an classification model made available to the classification unit comprising at least the first class of interest,
a feedback functionality for providing, in parallel with the acquisition and classification of the measurement data, feedback data based on user feedback by processing the assignment by a feedback procedure comprising:
verification information for the assignment of the measurement object to the first class of interest,
a definition of a new class of interest, and a first selection of the measurement data to be used for the classification,
wherein the feedback data is provided to a training unit comprising a training procedure, the training procedure being based on a machine learning algorithm and providing update information for the classification model, wherein the training procedure is adapted for processing feedback data from a multitude of feedback procedures.
9. The system according to claim 8 , wherein the feedback data being provided is based on at least one of:
explicit feedback by means of the feedback functionality for the acquisition of the measurement data or by a software used for processing of the measurement data, and implicit feedback.
10. The system according to claim 8 , wherein the feedback functionality is supported by a notification functionality providing a status information of the classification, and wherein the notification functionality is based on at least one of:
a visual representation of the status information,
an acoustic notification of the status information.
11. The system according to claim 8 , wherein the training procedure is adapted for at least one of:
processing feedback data provided by classifications of a multitude of measurement data, and
processing additional data provided to the training procedure.
12. The system according to claim 8 , wherein the update information for the classification model is made available to a multitude of classification algorithms.
13. The system according to claim 8 , wherein the classification model is provided out of a set of different models, differing from each other by at least one of:
a region-specific classification parameter,
a region-specific class of interest,
a time-specific classification parameter,
a time-specific class of interest,
an application-specific classification parameter, and
an application-specific class of interest,
wherein the classification model is provided based on at least one of a time information, a location information, and an application information, corresponding with the acquisition of the measurement data.
14. The system according to claim 8 , wherein the classification is based on at least one of:
a class of interest based on a semantic property,
a class of interest based on a geometrical property,
linear classification,
a support vector machine,
a quadratic classifier,
Kernel estimation,
boosting,
a decision tree,
deep learning, and
learning vector quantization.
15. A system for surveying for assigning a class of interest within measurement data, the system comprising:
a measurement device with a data acquisition unit for acquiring measurement data for at least one measurement object, which is at least partly captured by the measurement data, and
a classification unit for executing a classification of the measurement data by a classification algorithm for providing an assignment of the at least one measurement object to a first class of interest within the measurement data, based on:
the measurement data, and
an classification model made available to the classification unit comprising at least the first class of interest, the classification model having a region-specific and/or a time-specific classification parameter,
a feedback functionality for providing, in parallel with the acquisition and classification of the measurement data, feedback data based on user feedback by processing the assignment by a feedback procedure comprising at least one of:
verification information for the assignment of the measurement object to the first class of interest,
change information of the assignment of the measurement object to the first class of interest,
a definition of a new class of interest,
an instruction for removal of a class of interest from the classification model,
a first selection of the measurement data to be used for the classification, and
identification of a second selection of the measurement data to be ignored for further processing,
wherein the feedback data is provided to a training unit comprising a training procedure, the training procedure being based on a machine learning algorithm and providing update information for the classification model, wherein the training procedure is adapted for processing feedback data from a multitude of feedback procedures, namely feedback data taking into account a time of day, daytime and night time measurements, a particular part of the year, a season, a global position, longitude and latitude, or a climate zone.
16. The system according to claim 15 , wherein the feedback data being provided is based on at least one of:
explicit feedback by means of the feedback functionality for the acquisition of the measurement data or by a software used for processing of the measurement data, and implicit feedback.
17. The system according to claim 15 , wherein the feedback functionality is supported by a notification functionality providing a status information of the classification, and wherein the notification functionality is based on at least one of:
a visual representation of the status information,
an acoustic notification of the status information.
18. The system according to claim 15 , wherein the training procedure is adapted for at least one of:
processing feedback data provided by classifications of a multitude of measurement data, and
processing additional data provided to the training procedure.
19. The system according to claim 15 , wherein the update information for the classification model is made available to a multitude of classification algorithms.
20. The system according to claim 15 , wherein the classification model is provided out of a set of different models, differing from each other by at least one of:
a region-specific classification parameter,
a region-specific class of interest,
a time-specific classification parameter,
a time-specific class of interest,
an application-specific classification parameter, and
an application-specific class of interest,
wherein the classification model is provided based on at least one of a time information, a location information, and an application information, corresponding with the acquisition of the measurement data.
21. The system according to claim 15 , wherein the classification is based on at least one of:
a class of interest based on a semantic property,
a class of interest based on a geometrical property,
linear classification,
a support vector machine,
a quadratic classifier,
Kernel estimation,
boosting,
a decision tree,
deep learning, and
learning vector quantization.